Background: Artificial intelligence (AI) has the potential to enhance clinical decision-making, including in infectious diseases. By improving antimicrobial resistance prediction and optimizing antibiotic prescriptions, these technologies may support treatment strategies and address critical gaps in healthcare. This study evaluates the effectiveness of AI in guiding appropriate antibiotic prescriptions for infectious diseases through a systematic literature review. Methods: We conducted a systematic review of studies evaluating AI (machine learning or large language models) used for guidance on prescribing appropriate antibiotics in infectious disease cases. Searches were performed in PubMed, CINAHL, Embase, Scopus, Web of Science, and Google Scholar for articles published up to October 25, 2024. Inclusion criteria focused on studies assessing the performance of AI in clinical practice, with outcomes related to antimicrobial management and decision-making. Results: Seventeen studies used machine learning as part of clinical decision support systems (CDSS). They improved prediction of antimicrobial resistance and optimized antimicrobial use. Six studies focused on large language models to guide antimicrobial therapy; they had higher prescribing error rates, patient safety risks, and needed precise prompts to ensure accurate responses. Conclusions: AI, particularly machine learning integrated into CDSS, holds promise in enhancing clinical decision-making and improving antimicrobial management. However, large language models currently lack the reliability required for complex clinical applications. The indispensable role of infectious disease specialists remains critical for ensuring accurate, personalized, and safe treatment strategies. Rigorous validation and regular updates are essential before the successful integration of AI into clinical practice.
Background: Mold plasma cell-free DNA (cfDNA) polymerase chain reaction (PCR) is a promising noninvasive diagnostic modality for early diagnosis of invasive mold disease (IMD) in immunocompromised patients. Although mold cfDNA PCR has been shown to be highly accurate, the value of invasive procedures to collect specimens for conventional fungal diagnostics following plasma cfDNA testing remains unclear. Methods: This retrospective single-center cohort study included patients with mold plasma cfDNA PCR performed 7 days before or 2 days after invasive specimen collection. Mold PCR detected Aspergillus species, Mucorales agents, Fusarium species, and Scedosporium species. Invasive procedures included tissue biopsy and bronchoscopy. The primary endpoint was the concordance of mold plasma cfDNA PCR results with results of conventional fungal tests performed on tissue and bronchoalveolar lavage fluid. Results: Five hundred and six patients with mold plasma cfDNA PCR resulting ahead of invasive specimen (123 tissue and 426 bronchoalveolar lavage fluid) results were included, and 437 (86.4%) were immunocompromised. After adjudicating discordant results based on the European Organization for Research and Treatment of Cancer and Mycoses Study Group Education and Research Consortium definitions for IMD, mold plasma cfDNA PCR and invasive test results were 88.5% (448/506) concordant. In proven cases, 64.7% (11/17) of negative mold plasma cfDNA PCR results occurred in patients with fungal sinusitis (8) and limb infection (3). Nonhematologic malignancy and nonneutropenic state were statistically associated with negative mold plasma cfDNA PCR in patients with proven or probable IMD. Conclusions: Noninvasive mold plasma cfDNA PCR results were highly concordant with invasive specimen fungal test results, indicating risk-prone invasive specimen collection can be safely curtailed in immunocompromised patients with suspected IMD.
Background: Cryptococcosis is an increasingly common infection given the growing immunocompromised population worldwide. Cryptococcal antigen (CrAg) testing demonstrates excellent sensitivity and specificity and is the mainstay of diagnosis. However, there may be rare instances in which false-negative CrAg results can delay diagnosis and early treatment, which are critical to ensure positive outcomes. Case presentation: A 31-year-old man living with HIV/AIDS who was not taking antiretroviral therapy was hospitalized with fever, diarrhea, and headaches. CD4 count on presentation was 71 cells/uL, and HIV viral load was 3,194,949 copies/mL. Serum CrAg testing was initially negative, however CSF CrAg performed several days later was positive at 1:40 and blood and CSF cultures grew Cryptococcus neoformans. Colonoscopy revealed mucosal papules throughout the sigmoid colon, and tissue biopsy showed yeast within the lamina propria consistent with GI cryptococcosis. Given the high burden of disease, the original serum CrAg specimen was serially diluted and subsequently found to be positive at 1:2,560, confirming the postzone phenomenon. Conclusion: Cryptococcosis has a wide array of presentations including intraluminal GI disease, as seen in this patient. While serum CrAg testing displays excellent test characteristics, it is important for clinicians to be aware of the rare instances in which false-negative results may occur in the presence of excess antigen, as in this case.
Objectives: Bloodstream infection is associated with high mortality rates in critically ill patients but is difficult to identify clinically. This results in frequent blood culture testing, exposing patients to additional costs as well as the potential harms of unnecessary antibiotics. The purpose of this study was to assess whether the analysis of bedside physiologic monitoring data could accurately describe a pathophysiologic signature of bloodstream infection in patients admitted to the ICU. Design: Development of a statistical model using physiologic data from a retrospective observational cohort. Setting: University of Virginia Medical Center (Charlottesville, VA), a tertiary-care academic medical center. Patients: Critically ill patients consecutively admitted to either the medical or surgical/trauma ICUs with available physiologic monitoring data between February 2011 and June 2015. Interventions: None. Measurements and Main Results: We analyzed 9,954 ICU admissions with 144 patient-years of vital sign and electrocardiography waveform data, totaling 1.3 million hourly measurements. There were 15,577 blood culture instances, with 1,184 instances of bloodstream infection (8%). The multivariate pathophysiologic signature of bloodstream infection was characterized by abnormalities in 15 different physiologic features. The cross-validated area under the receiver operating characteristic curve was 0.78 (95% CI, 0.69–0.85). We also identified distinct signatures of Gram-negative and fungal bloodstream infections, but not Gram-positive bloodstream infection. Conclusions: Signatures of bloodstream infection can be identified in the routine physiologic monitoring data of critically ill adults. This may assist in identifying infected patients, maximizing diagnostic stewardship, and measuring the effect of new therapeutic modalities for sepsis.
25 September 2023 in Infection Control & Hospital Epidemiology
We used a strand-specific RT-qPCR to evaluate viral replication as a surrogate for infectiousness among 242 asymptomatic inpatients with a positive severe acute respiratory coronavirus virus 2 (SARS-CoV-2) admission test. Only 21 patients (9%) had detectable SARS-CoV-2 minus-strand RNA. Because most patients were found to be noninfectious, our findings support the suspension of asymptomatic admission testing.
9 December 2025 in Antimicrobial Stewardship & Healthcare Epidemiology